This is a second revision of a transdisciplinary proposal focusing on imaging and biology of early murine mammary cancers. We will employ quantitative serial magnetic resonance imaging with accompanying histology, immunohistochemistry, and X-ray fluorescence microscopy to understand the biology and natural history of in situ mammary cancer at initiation and progression. In addition, we will identify MRI parameters that detect various stages of mammary cancer, predict the future risk of progression to invasive cancer, and detect response to anti-angiogenic therapy. We will study C3(1)- SV40-Tag ('SVT') and MMTV-PyMT ('PMT') transgenic mice that develop MIN (mammary intraepithelial neoplasia; an intraductal neoplasia similar to DCIS in humans) and eventually invasive mammary cancers with a histological progression that is similar to poorly differentiated (basal type) and moderately differentiated breast cancers, respectively. Our collaboration was the first to demonstrate high resolution non-invasive imaging of in situ murine mammary cancer. Therefore, we are uniquely positioned to perform these highly integrated studies that take advantage of our unique strengths in small animal MRI imaging and cancer biology. We propose the following specific aims, taking advantage of our high field small animal MRI scanner, X-ray fluorescence microscopy at the Argonne National Labs, and extensive mammary gland biology research using animal models at the Ben-May Department of Cancer Research. The Specific Aims of the research are: 1. Optimize functional and anatomic MRI of pre-cancerous and intraductal neoplastic changes. Develop accurate co-registration of MRI with histology, immunohistochemistry and X-ray fluorescence images. 2. Measure sensitivity and specificity of MRI for detection of early mammary cancers and pre-cancers using pathologic assessment as the gold standard. 3. Perform a serial imaging study of the natural history of mammary cancer in the SVT and PMT mice. Determine what percentage of early stage lesions detected by MRI progresses to the invasive stage. Identify MRI-detectable physiologic and anatomic markers that differentiate malignant from benign changes, and predict whether cancers will progress. At each time point some mice will be sacrificed so that MRI parameters can be correlated with histology, immunohistochemistry, and X-ray fluorescence images. 4. Use serial MRI to evaluate response of early cancers to B20 (a bevacizumab-type monoclonal antibody that binds mouse VEGF) in the SVT mice. At each time point studied by MR, several mice will be sacrificed so that MRI findings can be correlated with biomarkers. This research develops a new paradigm for the study of early breast cancer - a transdisciplinary approach involving imaging and tumor biology - allowing us to improve breast cancer detection and control.